Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jan 18, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

1.0K

Optimizing Deep Learning-Based Crack Detection Using No-Reference Image Quality Assessment in a Mobile Tunnel

Chulhee Lee1, Donggyou Kim1, Dongku Kim1

  • 1Department of Geotechnical Engineering Research, Korea Institute of Civil Engineering and Building Technology (KICT), Goyang-Si 10223, Gyeonggi-Do, Republic of Korea.

Sensors (Basel, Switzerland)
|September 13, 2025
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Quality Assessment of High-Speed Motion Blur Images for Mobile Automated Tunnel Inspection.

Sensors (Basel, Switzerland)·2025
Same author

Unidirectional Real-Time Photoswitching of Diarylethene Molecular Monolayer Junctions with Multilayer Graphene Electrodes.

ACS applied materials & interfaces·2019
Same author

High-Yield Functional Molecular Electronic Devices.

ACS nano·2017
Same author

Correction to Three-Terminal Single-Molecule Junctions Formed by Mechanically Controllable Break Junctions with Side Gating.

Nano letters·2017
Same author

Gate-dependent asymmetric transport characteristics in pentacene barristors with graphene electrodes.

Nanotechnology·2016
Same author

Electrical characterization of benzenedithiolate molecular electronic devices with graphene electrodes on rigid and flexible substrates.

Nanotechnology·2016

Motion blur in mobile tunnel scanning systems significantly degrades deep learning crack detection. This study introduces a quality assurance framework using no-reference image quality assessment to improve model reliability for infrastructure maintenance.

Area of Science:

  • Computer Vision
  • Artificial Intelligence
  • Civil Engineering

Background:

  • Mobile tunnel scanning systems (MTSS) offer efficient inspection.
  • Motion blur (MB) from high speeds compromises deep learning crack detection accuracy.
  • Reliable crack detection is crucial for intelligent infrastructure maintenance.

Purpose of the Study:

  • Investigate the impact of horizontal MB on CNN-based crack detection in MTSS imagery.
  • Propose a data-centric quality assurance framework using no-reference image quality assessment (NR-IQA) to optimize model performance.
  • Identify effective NR-IQA metrics for assessing MB in MTSS data.

Main Methods:

  • Intentionally applied MB to public and real-world MTSS datasets.
  • Analyzed performance changes in ResNet, VGG, and AlexNet models.
Keywords:
convolutional neural networkcrack detectionmobile tunnel scanning systemmotion blurno-reference image quality assessment

More Related Videos

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.3K

Related Experiment Videos

Last Updated: Jan 18, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

1.0K
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.3K
  • Correlated four NR-IQA metrics (BRISQUE, NIQE, PIQE, CPBD) with crack detection F1 scores.
  • Developed a framework using PIQE and CPBD thresholds for image quality filtering.

    • Main Results:

    • ResNet34 F1 score dropped from 89.43% to 4.45% with increased MB intensity.
    • PIQE and CPBD showed strong correlations with F1 score (-0.87 and +0.82, respectively), indicating suitability for horizontal MB.
    • Filtering low-quality images using PIQE ≤ 20 and CPBD ≥ 0.8 improved AlexNet F1 score by 1.46%.

    Conclusions:

    • Horizontal MB significantly degrades CNN-based crack detection in MTSS.
    • PIQE and CPBD are effective NR-IQA metrics for assessing horizontal MB.
    • The proposed data-centric framework enhances MTSS data quality and optimizes deep learning models for reliable infrastructure maintenance.